Could you please help me understand the n_windows param from cross_validation with exogenous variables ? I went through the exogenous variables tutorial: https://github.com/Nixtla/mlforecast/blob/main/nbs/docs/how-to-guides/exogenous_features.ipynb My df_test looks exactly as the series_with_prices When I use n_windows=2 or some low number, everything works fine. I d like to simply go through the dataframe predicted 12 steps ahead at each timestep:

horizon = 12
n_windows = len(df_test) - horizon

# Perform cross-validation without refitting
cv_results = fcst.cross_validation(
    df_test,
    # static_features=[],
    n_windows=n_windows,
    h=horizon,
    step_size=1,
    refit=False,
    fitted=True
)

With the above setup however I get:

--> 462 self.fit_models(X, y)
    463 if fitted:
    464     fitted_values = self._compute_fitted_values(
    465         X_with_info=X_with_info,
...
--> 677         raise ValueError('Input data must be 2 dimensional and non empty.')
    679     # determine feature names
    680     if feature_name == 'auto':

ValueError: Input data must be 2 dimensional and non empty.

Is there something I am not getting ? : ) Thanks for patience.

Are all of your features static? If your dataframe looks like the example then price is dynamic

no all of my features are dynamic in fact. (At least if i understand dynamic correctly)

You should provide which columns are static (don't change over time for a single serie) through the

static_features

argument. By providing an empty list you're saying none are static (they all change over time) which is probably causing a bad join somewhere

They in fact do all change over time, so static_features = [] makes sense.

Oh I just saw it's commented out. What happens if you use it?

• I am trying with empty list and n_windows = 1000 and it seems to be working, • I just somehow need to set up n_windows to go through the entire test set and i am unsure how to do it.

1,000 windows seems excessive. Are you only forecasting one step ahead?

No 12 steps ahead, but I need to move through the entire test set in a walk forward manner. This I cant understand how to set up.

I think you can use the

input_size

but you have to be careful with the number of samples there, because if you're using lag5 for example and

dropna=True

then the features drop 5 rows, so you'll need

input_size=6

to get a single sample at the end

I see, perhaps I misunderstood, because the lags are present in the transformed dataframe, the "observing of data point" and the next 12 step prediction might not be what is really going on. • maybe I need to set up forecast horizon in the MLForecast.fit method

max_horizon: int, optional (default=None)
Train this many models, where each model will predict a specific horizon.

• it seems this is for something else. For when I assume different dynamics for each time horizon it seems.

Hmm not sure what you mean. If you set refit=True then the model is retrained on every step, if you set it to False it is only trained the first time. You can also set an integer, for example refit=7 retrains the model every 7 steps

aaa I see, amazing, amazing. that is exactly what I needed.